Something to consider: I just ran some installs on my current project to compare npm 5.0.3 with yarn 0.24.5 (both on Node 8.0.0):
Clean install (node_modules removed and cache cleaned, pre-existing package-lock.json/yarn.lock): No difference, each ~2m45s (2m39s - 2m56s, similar variance for both tools)
"Repeated" install (pre-existing node_modules and package-lock.json/yarn.lock): ~8s (npm) vs. ~1.5s (yarn)
It might not be that much of a gain to introduce a non-standard tool anymore. But I'll run some tests on Jenkins/Teamcity to be able to compare it to my initial tests last year: https://github.com/unic/estatico/issues/12#issue-186630282
Something to consider: I just ran some installs on my current project to compare npm 5.0.3 with yarn 0.24.5 (both on Node 8.0.0):
node_modules
removed and cache cleaned, pre-existingpackage-lock.json
/yarn.lock
): No difference, each ~2m45s (2m39s - 2m56s, similar variance for both tools)node_modules
andpackage-lock.json
/yarn.lock
): ~8s (npm) vs. ~1.5s (yarn)It might not be that much of a gain to introduce a non-standard tool anymore. But I'll run some tests on Jenkins/Teamcity to be able to compare it to my initial tests last year: https://github.com/unic/estatico/issues/12#issue-186630282